[Ffmpeg-devel-irc] ffmpeg-devel.log.20120314

[burek]

[03:54] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r29b0d94b43 10ffmpeg/libavcodec/dfa.c: 
[03:54] <CIA-17> ffmpeg: dfa: convert to bytestream2 API
[03:54] <CIA-17> ffmpeg: Protects from overreads.
[03:54] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[03:54] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[03:54] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r3aa661ec56 10ffmpeg/libavcodec/ (h264.c h264_ps.c): (log message trimmed)
[03:54] <CIA-17> ffmpeg: h264: improve parsing of broken AVC SPS
[03:54] <CIA-17> ffmpeg: Parsing the entire NAL as SPS fixes decoding of some AVC bitstreams
[03:54] <CIA-17> ffmpeg: with broken escaping. Since the size of the NAL unit is known and
[03:54] <CIA-17> ffmpeg: checked against the buffer end we can parse it entirely without buffer
[03:54] <CIA-17> ffmpeg: overreads.
[03:54] <CIA-17> ffmpeg: Fixes playback of
[03:54] <CIA-17> ffmpeg: 03Peter Ross 07master * r015da6e394 10ffmpeg/libavcodec/anm.c: 
[03:54] <CIA-17> ffmpeg: anm decoder: move buffer allocation from decode_init() to decode_frame()
[03:54] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[03:54] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[03:54] <CIA-17> ffmpeg: 03Paul B Mahol 07master * r6efe180782 10ffmpeg/tests/fate/wavpack.mak: 
[03:54] <CIA-17> ffmpeg: FATE: add shorthand to wavpack test
[03:54] <CIA-17> ffmpeg: Signed-off-by: Paul B Mahol <onemda at gmail.com>
[03:54] <CIA-17> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[03:54] <CIA-17> ffmpeg: 03Hendrik Leppkes 07master * rb2b0aa70ea 10ffmpeg/libavcodec/dxva2_vc1.c: 
[03:54] <CIA-17> ffmpeg: dxva2_vc1: fix decoding of BI frames
[03:54] <CIA-17> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[03:54] <CIA-17> ffmpeg: 03Hendrik Leppkes 07master * r7103c8350a 10ffmpeg/libavcodec/dxva2_vc1.c: 
[03:54] <CIA-17> ffmpeg: dxva2_vc1: pass the overlap flag to the decoder
[03:54] <CIA-17> ffmpeg: Signed-off-by: Anton Khirnov <anton at khirnov.net>
[04:39] <Compn> j-b : HV60 isom is really mpeg4 asp, and i think 'OggS' vorbis audio http://downloads.sourceforge.net/project/heroines/demo%20media/quicktime/fromspace_60.mov
[08:14] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r7230116000 10ffmpeg/ffmpeg.c: 
[08:14] <CIA-17> ffmpeg: ffmpeg: Fix planar audio input.
[08:14] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[08:18] <ubitux> beeh more complex stuff to deal with for the -af
[09:34] <j-b> michaelni: http://source.ffmpeg.org
[09:35] <av500> use the source, luke
[09:36] <j-b> use the luke, source
[10:05] <cbsrobot> in a 8 bit yuva image, the alpha channel is normally 0-255 ?
[13:42] <codevomit> hi, I was wondering what methods would I use to get a list of directshow devices
[16:59] <ubitux> is there differences between the two AAC iso standards? (ISO/IEC 13818-7 and ISO/IEC 14496-3?)
[16:59] <ubitux> or it's basically the same paper?
[17:01] <merbanan> mpeg-4 14496-3, mpeg-2 13818-7
[17:02] <merbanan> the mpeg-4 is the usable one
[17:03] <ubitux> what do you mean by usable?
[17:04] <ubitux> what's the point of this part 7 in the mpeg-2 if it's not "usable"?
[17:04] <av500> its a standard, its not meant to be usable :)
[17:05] <ubitux> printed on paper you can certainly make an awesome airplane with it, so please stop saying this
[17:06] <ubitux> i'm just curious about the differencies between the two
[17:06] <ubitux> even if in practice aac is not commonly (at all?) used in mpeg-2
[17:06] <merbanan> the mpeg2 doesn't have sbr/ps etc
[17:07] <av500> does it have different bitstream at least?
[17:12] <merbanan> who knows, it's not really used
[17:12] <gnafu> AAC Constrained Baseline Profile?
[17:12] <av500> Constipated
[17:16] <gnafu> "[...]strained" --> "Constipated"
[19:25] <Compn> i got a strange mail that i swear was spam
[19:25] <Compn> any ideas ?
[19:25] <Compn> Dear Sirs,
[19:25] <Compn> Let me introduce myself: I work in Ericsson Sourcing, had your email address in order to possibly adopt your Open Source SW into one of our Telecom products.
[19:25] <Compn> In order to do so, would be grateful if you could fill in the attached template with relevant Company information: where applicable.
[19:25] <Compn> This is standard process when we wish to acquire new suppliers: we are evaluating a support service from you.
[19:25] <Compn> why should i fill out a form so ericsson can use oss in its crappy phones ?
[19:28] <gnafu> Hey, at least they didn't ask you to send a several-thousand-dollar application fee to help them regain access to their bank account.
[19:31] <Compn> i'm thinking up a reply in my head
[19:31] <Compn> 'they pay you to look for emails, they can pay you to fill out forms too'
[19:32] <Compn> or would it make sense to fill out the form so ericsson includes mplayer/ffmpeg and then submits patches ?
[19:32] <Compn> hmmm
[19:40] <Compn> anyone want to fill out some form ?
[19:46] <pasteeater> make it a SoC project
[19:53] <Compn> i'd look at it but its .xls and i dont have such word processing programs on my box
[21:23] <pasteeater> Compn: i think google docs can import that junk. or forward it to me.
[21:24] <Compn> yeah google docs imported it
[21:24] <Compn> what email ?
[21:24] <Compn> i mean whats your email ?
[21:24] <Compn> its like 'who to contact at your company' 'who to pay money to' 'how much for your software' 
[21:25] <Compn> they sent it to projects at mplayerhq.hu , but i have no idea if thats for ffmpeg or mplayer (or rtmpdump haha)
[21:26] <Compn> we should put some contact info on the websites
[21:43] <Compn> Entries for the International Teletext Art Festival are currently being broadcast on teletext pages 525-545 of Finnish broadcaster YLE.
[21:50] <Daemon404> WTF
[21:50] <Daemon404> "A problem internal to GDB has been detected."
[21:50] <Daemon404> then it explodes
[21:52] Action: durandal_1707 does not have gdb installed
[21:52] <ohsix> ...
[21:52] <Daemon404> backtraces are kind of useful.
[21:53] Action: durandal_1707 prefers brains
[21:53] <Daemon404> i think you mean wasting time
[21:53] <Daemon404> tracking down what function it crashes in
[21:53] <Daemon404> line etc
[21:53] <ohsix> if it compiles, ship it!
[21:54] <durandal_1707> Daemon404: well if you are debuggind someone else code it may or may not help :)
[21:54] <ohsix> the philosophical question is, if you don't have a debugger, do you have bugs?
[21:55] <Daemon404> hmm
[21:55] <Daemon404> avformat_new_strea(s, NULL);
[21:55] <Daemon404> is crashing
[21:55] <Daemon404> stream*
[21:59] <ubitux> btw, no one interested in the timecode writing in mov?
[21:59] <durandal_1707> you mean writing integers?
[21:59] <ubitux> no, the tmcd track
[22:00] <ubitux> the patch i sent a few days ago
[22:00] <ubitux> kind of the last big brick in the timecode support
[22:00] <durandal_1707> ahh, there was one bug report...
[22:01] <Daemon404> ubitux, i thought ffmbc did this.
[22:02] <ubitux> there is an issue for the extraction, not sure about write support
[22:02] <ubitux> Daemon404: certainly, but with the patch ffmpeg should do it too now
[22:11] <CIA-17> ffmpeg: 03Aaron Colwell 07master * rbda5b6627a 10ffmpeg/libavformat/matroskaenc.c: 
[22:11] <CIA-17> ffmpeg: matroskaenc: change cluster_pos to use -1 as the "I'm not currently working on a cluster" signal instead of 0.
[22:11] <CIA-17> ffmpeg: This avoids problems
[22:11] <CIA-17> ffmpeg: where avio_tell() returns 0. I've updated all the checks against
[22:11] <CIA-17> ffmpeg: cluster_pos
[22:11] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:04] <TACPILOT> found some documentation referring to an issue with how libavcodec handles h.264 interlaced content. I dont know how old the info is? 
[23:04] <TACPILOT> http://pastebin.com/e6iTmFYY
[23:04] <TACPILOT> has this issue been resolved with current versions ??
[23:05] <TACPILOT> the (odd and even) jumping around is of concern to me
[23:06] <TACPILOT> oops wrong link this is valid
[23:06] <TACPILOT> http://pastebin.com/Y0UjZm9D
[23:09] <CIA-17> ffmpeg: 03Baptiste Coudurier 07master * rfa35ade8f9 10ffmpeg/libavfilter/libmpcodecs/vf_fspp.c: vf_fspp: fix compilation with llvm
[23:10] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * rf43a16049e 10ffmpeg/cmdutils.c: (log message trimmed)
[23:10] <CIA-17> ffmpeg: cmdutils: Fix cpuflags so the flags dont pull in and out other flags.
[23:10] <CIA-17> ffmpeg: Note, this doesnt break compatibility with libav, as libav
[23:10] <CIA-17> ffmpeg: has implemented a incompatible and more limited system under the same
[23:10] <CIA-17> ffmpeg: -cpuflags command line option we used since some time.
[23:10] <CIA-17> ffmpeg: The differences to libav for example are we can do things like
[23:10] <CIA-17> ffmpeg: ffmpeg -cpuflags -sse+mmx -cpuflags +3dnow
[23:10] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r9e2ee46206 10ffmpeg/libavfilter/libmpcodecs/vf_fspp.c: 
[23:10] <CIA-17> ffmpeg: vf_fspp: add ff_ prefix to non static variables.
[23:10] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:10] <CIA-17> ffmpeg: 03Michael Niedermayer 07master * r3bd749f335 10ffmpeg/doc/avtools-common-opts.texi: 
[23:10] <CIA-17> ffmpeg: docs: Add examples for -cpuflags
[23:10] <CIA-17> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:13] <michaelni> TACPILOT, all known bugs are at our bugtracker, ffmpeg.org/trac
[23:13] <michaelni> if its listed there its known, if its not listed there its not known
[23:14] <michaelni> if its not listed there AND you can reproduce an issue with latest ffmpeg git then please report it
[23:15] <TACPILOT> will take a look TY
[23:18] <Daemon404> http://i.imgur.com/BXFxQ.png
[23:18] <Daemon404> ^ does this look like anything to anyone
[23:19] <Plorkyeran> looks like a penis
[23:20] <Daemon404> of course.
[23:20] <durandal_1707> Daemon404: what is that from?
[23:21] <Daemon404> a half-reverse-engineere'd codec
[23:21] <durandal_1707> name of codec?
[23:21] <Daemon404> this eitehr has a 2nd part to entropy decoding, or a really strange form of prediciton
[23:21] <Daemon404> zerocdec
[23:21] <Daemon404> i dunon why you keep asking the names
[23:21] <Daemon404> as its not really relevant
[23:22] <durandal_1707> looks like wavelet to me
[23:22] <Daemon404> i somewhat doubt this.
[23:30] <gnafu> What's it /supposed/ to look like?
[23:32] <Daemon404> like that obviously, since it's not fully decoded
[23:32] <Daemon404> :P
[23:32] <floater> hey who can see my shadow? i am the amoeba like thing in your eye
[23:44] <Compn> floater : what 
[23:44] <Compn> Daemon404 : we ask so we know if you are RE'ing a raw codec, a dct based codec, or wavelet codec
[23:44] <Compn> because half-decoded images look different for each of those
[23:45] <floater> Compn << http://en.wikipedia.org/wiki/Floater :)
[23:45] <Daemon404> Compn, "lossless"
[23:45] <Daemon404> with zlib entropy encoding
[23:46] <Compn> no clue
[00:00] --- Thu Mar 15 2012